Chapter 19: Flirtation as Social Performance — Scripts, Improvisation, and Ambiguity

Dr. Adaeze Okafor and Dr. Carlos Reyes had been arguing productively about flirtation for three years by the time they sat down to compare behavioral coding data from their field teams in six countries. Okafor had always suspected that what Reyes called "courtship solicitation behaviors" was actually a culturally specific repertoire being dressed up in evolutionary language. Reyes was willing to concede that specific gestures varied — but insisted that the function of those gestures, and the underlying sequence they formed, were recognizable across all twelve of the project's countries.

The behavioral coding data from Year 3 of the Global Attraction Project gave both of them something to work with. The Okafor-Reyes flirtation coding protocol had been developed through painstaking iteration — the initial behavioral catalog developed by Reyes's Buenos Aires team had to be substantially revised when Okafor's Nigerian fieldworkers reported that several behaviors simply didn't occur in the Lagos bar contexts they were observing, while other behaviors that were frequent there didn't appear in the original catalog at all. What emerged from the reconciliation was both more complicated and more interesting than either team had expected.

Some things were consistent across Nigeria, South Africa, Sweden, Germany, Japan, and Brazil: behavioral synchrony increased in flirtatious interaction relative to neutral interaction. Smiling — not any particular kind of smile, but elevated smiling generally — was universal. And in all six cultural contexts, there was what Okafor's team called "structured ambiguity": flirtatious behavior was characterized by a distinctive combination of signals that could be interpreted either as interested or as merely friendly. This was not a failure of the behavior to clearly communicate. It was, they concluded, the behavior's central feature.

19.1 What Is Flirtation? A Working Definition

Defining flirtation precisely turns out to be surprisingly tricky. Everyone recognizes it — and considerable disagreement exists about where it begins and ends. The French distinction between flirter and séduire (roughly, playful signaling versus sustained romantic pursuit) captures something the English word "flirtation" partially collapses. In everyday language, "flirting" describes behaviors ranging from a momentary provocative glance to extended mutually pleasurable courtship play that stops just short of explicit romantic declaration.

For our purposes, a useful working definition comes from the sociologist David Givens and, separately, from Timothy Perper's observational work: flirtation is the use of indirect, playful, and deniable signals to communicate potential romantic or sexual interest, while maintaining the interactional frame of ordinary friendly sociability.

Several elements of this definition are worth unpacking:

Indirect: Flirtation does not typically say what it means directly. It operates in suggestion, implication, double entendre, and gesture rather than direct declaration. This is not a design flaw — it is, as we will argue at length in Section 19.7, the adaptation's core function.

Playful: Flirtation is typically conducted in a non-serious frame — a context of play that differs from the earnest seriousness of direct romantic disclosure. Laughter, teasing, wit, and lightness are hallmarks of flirtatious interaction.

Deniable: A flirtatious signal can, when challenged, be plausibly reinterpreted as something less charged. "I was just being friendly." This deniability is not incidental to flirtation — it is structural.

Within a friendly-sociability frame: Flirtation occurs inside ordinary social interaction, not as a separate announced activity. It uses the conventions of friendly conversation as its cover and medium.

💡 Key Insight: Flirtation is not a failed or ambiguous version of romantic declaration. It is a distinct social behavior with its own structural logic. Understanding it requires accepting that its defining features — indirectness, playfulness, and deniability — are functional, not accidental.

The problem of subjective versus behavioral definitions: One of the conceptual challenges in flirtation research is that the behavioral definition (what someone does) and the subjective definition (what they intend or experience) do not always match. A person may perform behaviors that are coded as flirtatious by an outside observer without intending to flirt — they may simply be warm, tactile, and animated in their social style. Conversely, a person may intend to flirt while performing behaviors that are ambiguous enough that an observer would not code them as flirtatious.

Researchers have addressed this through what is called the triangulation strategy: combining behavioral coding (what happens), self-report (what the person intended), and peer perception (how observers read the interaction). When all three align, we have the clearest cases of flirtation. When they diverge, we have the cases that are most theoretically interesting — and most practically consequential, since it is the divergent cases that generate misunderstanding.

The Global Attraction Project used a version of this triangulation approach in its Year 3 behavioral coding protocol. Field observers coded behaviors; participants in some sub-studies completed brief post-interaction questionnaires about their intentions; and cultural informants reviewed ambiguous cases. This multi-method approach is methodologically stronger than any single data source, though it also introduces complexity in comparing data across sites with different sub-study designs.

19.2 The Dramaturgical Lens: Goffman and the Performance of Self

Erving Goffman's dramaturgical framework, developed primarily in The Presentation of Self in Everyday Life (1959), offers a powerful conceptual lens for analyzing flirtation. Goffman proposed that social interaction is fundamentally theatrical: people are always performing for audiences, managing impressions, and navigating the implicit scripts of social situations.

In Goffman's framework, every interaction involves front stage behavior (the managed, public presentation) and back stage reality (the private, unmanaged self). Social life is organized by frames — shared understandings of what kind of situation we are in and what kind of behavior is therefore appropriate. And effective social actors deploy face work — the management of public presentations to protect both their own dignity and others'.

All of these concepts illuminate flirtation with particular sharpness.

Front stage/back stage: Flirtation is front stage behavior that carries back stage information. The person flirting is performing interest in a way that is managed and deniable — they are putting forward a signal that could be received as interest but could also be claimed as mere friendliness. This double-front-stage quality is specific to flirtation and distinguishes it from other forms of courtship communication.

Frame: Flirtation is possible because of the existence of a shared "friendly sociability" frame within which more charged signals can be inserted and withdrawn. The flirtatious signal operates by briefly stretching the frame — introducing a moment of ambiguity about whether we are in a purely social situation or something more — and then allowing the frame to return to social normality. This is the mechanism by which flirtation produces what Okafor's team would call its structured ambiguity.

Face work: Flirtation is exquisitely face-protective. Direct romantic declaration carries enormous face risk — rejection is total and public. Flirtatious ambiguity allows both parties to maintain face regardless of outcome: if the signal is not received or reciprocated, both parties can pretend it was never sent.

🔗 Connections: Goffman's framework connects to the self-presentation research we examined in Chapter 15, which found that people calibrate their self-presentation based on identity concerns and audience awareness. Flirtation is a specific case of self-presentation under particularly high uncertainty and face risk.

The dramaturgical perspective also illuminates something that purely psychological accounts of flirtation miss: flirtation is inherently interactional, not individual. It is not something you do — it is something you and another person create together. Goffman would say that the flirtation exists in the interaction, not in the intentions of either participant. This is why flirtation can be performed mutually and enjoyably without either party being "serious," and why it can be performed seriously while appearing playful.

19.3 Flirtation Scripts: Culturally Prescribed Sequences

A script in social psychology refers to a shared cognitive template for how a sequence of social events typically unfolds. Scripts are not rigid prescriptions — they are general outlines that organize behavior and create mutual expectations. People follow social scripts partly because deviation from them is confusing or requires explanation, and partly because scripts coordinate behavior between people who don't know each other well.

Courtship scripting was formally theorized by Gagnon and Simon in their sexual script theory (1973), which argued that sexual and romantic behavior is organized by culturally transmitted scripts at three levels: the cultural (shared norms about how courtship proceeds), the interpersonal (the negotiated script two people construct between them), and the intrapsychic (the personal fantasies and desires that motivate individual behavior). These three levels often align imperfectly, and the misalignments between them are a source of much courtship misunderstanding.

Flirtation scripting is real and empirically demonstrable. Research on courtship interaction in naturalistic settings consistently finds that flirtatious interactions unfold in recognizable sequences that vary across cultures in their specific content but share general structural features. The first person to document this rigorously, through systematic behavioral observation, was Timothy Perper.

Scripts and improvisation: The relationship between script and improvisation in flirtation is worth dwelling on. Calling flirtation "scripted" might seem to imply that it is mechanical or predetermined — that people are simply running a program. This is not right. Think of a script less like a screenplay (fixed dialogue, no deviation) and more like a jazz standard: there is a shared structure — a chord progression, a familiar melodic line — that both musicians know, which allows them to improvise freely within it. The script provides a framework that makes coordination possible; the improvisation within that framework is where the individual personality, chemistry, and situational specifics appear.

This analogy captures something important about why flirtation feels both familiar and unique in each instance. You recognize what is happening — it follows a pattern you know — and yet each interaction is genuinely different, shaped by the specific people, the specific setting, and the specific energy of the moment. Skilled flirters are not people who have memorized a better script; they are people who are more fluent in the underlying structure, and therefore more free to improvise within it.

Script violations: What happens when someone violates the script? The consequences depend on the direction of violation. Moving too fast — escalating before the other party has had time to establish readiness — is experienced as jarring and presumptuous. Moving too slow — failing to escalate when the other party has clearly signaled readiness — is experienced as obtuse or as a rejection in disguise. The script establishes a pace expectation, and deviations from it in either direction carry costs. Understanding this is one reason why "reading the room" matters so much in flirtatious interaction — it is partly a matter of reading where in the script the other person thinks you are, and calibrating accordingly.

19.4 The Perper-Moore Flirtation Sequence

Timothy Perper's observational work, conducted primarily in bars and social settings in the late 1970s and 1980s and synthesized in Sex Signals: The Biology of Love (1985), identified what has become the most cited description of courtship sequence in naturalistic settings: a progressive behavioral escalation in which each step, if reciprocated, leads to the next.

Perper's sequence, subsequently refined and complemented by Monica Moore's parallel observational work (discussed in Case Study 1 of Chapter 19), involves four broad phases:

Phase 1 — Approach/Positioning: One party (in Perper's predominantly heterosexual sample, more often the woman) positions herself in the other's visual field, or engages in behaviors (elevated grooming, increased laughter, postural signals) that make herself available to be approached. This is not passivity — it is active invitation-making.

Phase 2 — Turn (Initial Engagement): The two parties begin interaction. Conversation starts, often with neutral or mundane topic content. The flirtatious signals are not necessarily in the words at this stage but in the paralinguistic and nonverbal channels: tone, eye contact, the willingness to prolong the interaction beyond what the ostensible topic warrants.

Phase 3 — Close (Progressive Reduction of Distance): Physical and conversational distance progressively reduces. Each small reduction — leaning slightly closer, turning more directly toward the other, touching briefly — is either accepted or not accepted. Acceptance invites further reduction. Non-acceptance does not necessarily end the interaction but signals a limit.

Phase 4 — Touch and Escalation: Touch enters the sequence, beginning with low-intimacy touch (arm, shoulder) and potentially escalating toward more intimate touch if each stage is reciprocated. Touch, in Perper's account, is the clearest moment of explicit mutual acknowledgment that the interaction has moved out of purely social territory.

📊 Research Spotlight: What made Perper's work genuinely scientific rather than merely observational was his systematic coding methodology. He observed hundreds of interactions and coded behavioral sequences, allowing him to identify which specific sequences predicted successful courtship escalation (both parties continuing through further phases) and which were associated with stalling or discontinuation. The predictive validity of the sequence approach — where prior-phase reciprocation predicts next-phase reciprocation — is its most interesting feature.

The critical role of reciprocation: The most theoretically important element of the Perper model is that it is not a unilateral script — it is a mutually negotiated sequence in which each step requires the other party's implicit consent through behavioral response. Failure to reciprocate at any stage functionally terminates the sequence, though both parties can continue to interact normally in the social frame. This structure makes the sequence both ethically important (it operationalizes ongoing consent) and psychologically complex (failures of reciprocation are sometimes missed or misinterpreted).

19.5 Female-Initiated Courtship Signals: The Ethological Literature

One of the most important and underappreciated findings in the behavioral literature on courtship is that women are typically the initiators of the early phases of heterosexual courtship — not men. This finding, which runs counter to the dominant cultural script in many Western contexts (where men are supposed to "make the move"), emerges consistently from naturalistic observation.

Monica Moore's 1985 ethological study of women's courtship behavior in bar settings documented 52 distinct non-contact solicitation behaviors that women performed in the vicinity of men they apparently found attractive, and correlated these with the probability of a man approaching. The behaviors ranged from room-encompassing glances (scanning the environment while remaining aware of a target) to neck presentations, hair flips, giggling at the target, and the "flirting glance" (short darting glance followed by gaze aversion). Moore found that the rate of these behaviors was strongly predictive of whether a man would approach — and that women who did not perform them were rarely approached regardless of their physical attractiveness.

This finding has been partially replicated in subsequent studies (Case Study 1 examines the Moore study in detail) and interpreted in two compatible frameworks:

The evolutionary interpretation: Women as the higher-investing sex (in a Trivers parental investment framework) face greater costs from mating with unsuitable partners, and therefore have evolved to be more selective in whom they "allow" to approach. The female courtship signal functions as a green light that reduces the woman's face risk from approach while allowing her to select which men she invites.

The social constructionist interpretation: In contexts where direct female approach of a male is socially penalized (seen as too forward, unfeminine, or aggressive), signaling through indirect invitation-making is an adaptive response to social norms, not an evolved program. The behavior is real; its cause is an open question.

As with most debates in evolutionary social psychology, both frameworks capture something true. The behavior is documented; the mechanism is contested.

The Grammer extension: Karl Grammer's parallel ethological research, conducted in Austrian discotheques and published in 1990, replicated and extended Moore's findings. Grammer's team coded both female and male behavior and found a sequential relationship: female solicitation behaviors predicted male approach at a statistically significant rate, and male approach behavior was itself predictive of further female response (either increased or decreased solicitation behavior depending on whether the approach was welcomed). This bidirectional coding represented an advance over Moore's single-sex focus: it demonstrated that the courtship interaction was genuinely sequential and mutually regulated, not simply a female performance followed by a male action.

Grammer also noted that the behavioral repertoire differed somewhat between the Austrian sample and the North American patterns Moore had described — confirming that while the structural sequence was similar, the specific behaviors were culturally inflected. This early cross-cultural comparison prefigured the more systematic work that the Okafor-Reyes protocol would attempt to accomplish decades later.

What the finding does not mean: It is important to be clear about the interpretive limits of these findings. The observation that women's solicitation behaviors precede and predict male approaches in bar settings does not mean:

• That women are "responsible" for romantic outcomes in the sense of bearing moral responsibility
• That men who do not receive solicitation signals are entitled to approach anyway
• That this pattern describes all courtship in all contexts
• That individuals within the population must conform to the statistical average

It means that in naturalistic observations of heterosexual courtship in Western bar settings, a particular sequential structure has been observed with statistical regularity. That structure is interesting precisely because it challenges the cultural narrative of male initiation — but it should not be replaced by a narrative that simply relocates the initiation burden to women.

19.6 Male Courtship Display Behaviors

If women's courtship solicitation signals function as invitations to approach, men's courtship display behaviors function primarily in the approach and status-signaling phases. Research has documented several male courtship display patterns, though the literature here is somewhat less systematic than Moore's work on female solicitation.

Resource display and space-maximizing behavior: Men in courtship contexts have been observed to use more physical space (spreading posture, occupying more of a table or couch), to engage in more dominant and expansive body language, and to make displays relevant to social status and competence. This connects to the large literature on resource signals and mate value we examined in Chapter 8.

Vocal display: Men lower their voices and speak more slowly in courtship contexts. Research by Singh and colleagues found that observers reliably rate men's voices as more attractive when they are modulated in these ways, and that men do this to some degree automatically when talking to attractive women.

Humor production: Men in courtship contexts produce more humor attempts, and research finds that men rate their own humor production as higher during flirtatious interaction. Whether they are actually funnier is a separate question — what is clear is that humor production functions as a display behavior for men in many cultural contexts.

The approach itself: In Perper's and Moore's accounts, the man's approach following the woman's solicitation signal is itself a display — it demonstrates confidence and social competence. Men who fail to approach despite receiving signals are not merely shy — they may be misreading the signals, or facing a social norm that makes approach feel high-risk regardless of signaling.

⚖️ Debate Point: The display-signaling model of male and female courtship roles has been criticized for being too tightly organized around a heterosexual, male-approach/female-invite binary. Same-sex courtship research (of which there is considerably less) suggests somewhat different patterns, and even within heterosexual courtship the model is a statistical tendency rather than a rule. Women approach; men signal; and in both cases individual variation and social context shape behavior at least as much as any average tendency.

19.7 The Ambiguity Function: Why Flirtation Is Strategically Unclear

Here we arrive at what is, theoretically, the most interesting section of this chapter: the question of why flirtation is built to be ambiguous. This is not an obvious question — we might expect that a communication whose purpose is to signal interest would be designed to maximize clarity, not minimize it. The fact that flirtation systematically obscures rather than reveals its intent demands an explanation.

Several explanations have been offered, and they are not mutually exclusive:

The face-protection hypothesis: The most widely cited explanation, advanced by Goffman's framework and extended by subsequent researchers, is that ambiguity protects both parties from the face costs of explicit rejection. When a romantic declaration is made clearly and directly, rejection is also clear and direct — and carries significant costs: embarrassment for the initiator, an awkward ongoing relationship for both, and potential damage to the social network in which both are embedded. Flirtation's ambiguity allows a person to signal interest at low cost: if the signal is reciprocated, escalation can follow; if it is not, neither party need acknowledge that a signal was ever sent.

This explanation is strongly supported by the consistency of flirtation's deniability property across cultures. In every context where flirtation has been studied, the behaviors used are ones that could plausibly have alternative meanings. This appears to be a structural requirement, not a cultural accident.

The quality-testing hypothesis: An alternative explanation, proposed from an evolutionary perspective, is that ambiguity functions as a test of the potential partner's social intelligence and attentiveness. Someone who can detect a subtle, deniable signal of interest — who is socially perceptive enough to notice what is not being said explicitly — may thereby reveal something valuable about their cognitive and social competence. The difficulty of the signal functions as a filter.

This hypothesis is harder to test directly, but it is consistent with findings suggesting that people who are more socially intelligent are better at reading flirtatious signals, and that this ability is itself viewed as attractive.

The arousal and suspension function: A third possibility, emphasized in literary and philosophical accounts of courtship and given some psychological grounding in research on reward timing, is that ambiguity is pleasurable. The sustained uncertainty of a flirtatious exchange — not knowing whether the interest is returned, whether this is "real" or playful — creates a state of heightened attention and arousal that is intrinsically rewarding. Ambiguity keeps the game alive.

This is the basis for the distinction between terminal flirtation (which is the opening phase of courtship that will eventually resolve into explicit romantic engagement) and recreational flirtation (which is pleasurable mutual play that is not intended to lead to anything further). Research suggests that both forms are common, and that one of the more frequent misreadings in flirtatious interaction is one party treating recreational flirtation as terminal while the other intends it as recreational.

The strategic coordination hypothesis: A final perspective, drawn from game theory, frames flirtation as a mutual coordination game under uncertainty. Neither party knows for certain how interested the other is; both parties are trying to assess mutual interest while managing costs. Flirtation's sequence structure — with its graduated escalation and reciprocation requirements — allows for low-cost early signaling and information gathering before either party commits to an explicit declaration.

💡 Key Insight: The function of ambiguity in flirtation is not a communication failure — it is a communication strategy. It serves multiple parties' interests simultaneously: it protects face, extends pleasurable tension, tests social intelligence, and allows for low-cost early information gathering. When we understand this, "why didn't they just say what they meant?" stops being the right question. The indirectness is what they meant.

Okafor's fieldwork in Nigeria and South Africa documented that the specific content of flirtatious behaviors varied considerably between the two contexts — the gesture vocabulary, the norms around approach, the degree of verbal versus nonverbal signaling — but that this structural property of deniability was consistent. Whether you were in a Lagos nightclub or a Johannesburg university common room, flirtatious interaction was characterized by signals that could be interpreted as mere friendliness. The mechanism was universal; the content was not.

19.7b The Game Theory of Flirtation: A Brief Note

For students familiar with basic game theory concepts, it is worth briefly noting that the strategic coordination hypothesis for flirtation ambiguity can be formally modeled as what game theorists call a signaling game. In a classic signaling game, a sender has private information (their actual level of interest or suitability) that the receiver wants to know, but cannot directly observe. The sender can send signals, but signals are only informative if they are costly enough that low-quality senders cannot profitably imitate high-quality senders.

The problem with applying standard costly signaling theory directly to flirtation is that flirtation signals are, in a sense, cheap — they are specifically designed to be deniable and low-cost to send. This seems to predict that they should be uninformative, because a disinterested person could send them just as easily as an interested one.

The resolution is that flirtation signals are not purely about the content of interest; they are about the initiation of a coordination process. The signal's value is not in its difficulty (as in classic costly signaling) but in its sequential structure: each reciprocated step commits the sender slightly more than the last, progressively raising the cost of further engagement for a disinterested party. The early, cheap signals are informationally weak; the later, more escalated signals (touch, sustained gaze, explicit verbal play) are more costly to perform without genuine interest. The sequence as a whole performs the function that costly signaling performs in other domains — filtering out disinterested partners — even though no single early signal is sufficient to do so.

19.8 Rejection Ambiguity: When "No" Is Unclear vs. When It Is Unmistakable

The same ambiguity that protects face and creates pleasurable uncertainty also creates a genuine problem: the interpretation of disinterest. If flirtatious signals are designed to be deniable, how is non-flirtatious response interpreted? When is "I'm not interested" clear?

The empirical research here is uncomfortable but important. Studies of how people interpret responses to flirtatious advances consistently find that some people — and the research suggests this is more common among men in heterosexual interaction, though not exclusive to them — show a systematic tendency to over-read interest in ambiguous situations. This is sometimes called the sexual overperception bias: the tendency to interpret ambiguous social cues as indicating more sexual or romantic interest than the other person intends.

Research by Haselton (2003) proposed an evolutionary explanation for this asymmetry: given the costs of missed opportunities versus the costs of false alarms, natural selection may have shaped men to err on the side of over-perceiving interest (you lose little by making an approach that fails) while shaping women to err on the side of under-perceiving interest (the cost of a sexual encounter with an unsuitable partner is higher). This is an evolutionary hypothesis, and like all such hypotheses it requires epistemic humility — it has some support but is not established fact.

What is clear from the empirical literature is that the over-perception bias is real as a population tendency, causes real harm (unwanted advances, harassment, misread boundaries), and is not a fixed biological trait but is modified by social context, norms, and education.

The question of when "no" is clear is also worth confronting directly. Research on refusal behavior finds that many rejections are indirect — people soften refusals with excuses, hedging, or ambiguous wording rather than direct refusal. This is partly the same face-protection logic that drives flirtation's ambiguity. But the consequence is that the ambiguity of flirtation can extend into rejection, creating genuine confusion about whether an advance is being declined or redirected.

The clearest rejections in behavioral research are: physical withdrawal, direct verbal refusal, absence of any reciprocation across multiple repeated escalation attempts, and ongoing deflection to group or casual interaction frames. But even some of these can be misconstrued by someone motivated to misconstrue them.

🔵 Ethical Lens: The ambiguity function of flirtation is not a license for willful misreading of rejection. The same social intelligence that allows people to read subtle signals of interest should — and often does — also allow reading of signals of disinterest. The research on over-perception bias describes a statistical tendency, not an excuse. Consent negotiation requires engaging with the signals that are actually being sent, not only the ones that are wished for.

19.9 Playfulness, Laughter, and the Non-Serious Frame

Laughter is among the most consistently documented features of flirtatious interaction. Couples who are flirting laugh more — both more genuine laughter and more responsive laughter (laughing because the other person said something, not necessarily because it was especially funny). This is not merely coincidental: laughter serves several specific functions in the courtship context.

Laughter signals benign intent. Robert Provine's research on laughter found that it occurs most frequently in social interaction rather than in response to jokes, and that its primary function is social bonding rather than humor response. In a flirtatious context, laughter signals comfort, approachability, and the absence of threat — all of which facilitate the risk-taking that flirtatious signaling requires.

Laughter maintains the non-serious frame. By laughing — by treating the interaction as light and playful — both parties signal that they are in a flirtatious, not a serious, mode. This non-serious frame is protective: if the signals do not resolve into romantic interest, both parties can point to the playfulness as evidence that "it was just fun, nothing serious." The frame does the same face-protection work as ambiguity, but in a different register.

Laughter coordinates arousal. Mutual laughter has been shown to synchronize physiological arousal between interaction partners. Shared laughter creates a moment of mutual experience that builds the sense of "we" — a small instance of the shared reality that is a building block of intimacy.

Teasing — a form of playful challenge that is characteristic of much flirtation — operates similarly. Teasing is the use of mild mockery that is clearly not mean-spirited, delivered within a shared non-serious frame. It accomplishes something interesting: it signals a degree of closeness (only close enough people can tease each other) while also displaying wit and social confidence. Research by Keltner and colleagues on teasing in close relationships found that teasing is a marker of established familiarity — and that in early interaction, flirtatious teasing serves to propose that familiarity rather than simply reflecting it.

The humor production asymmetry: Research on humor in courtship has documented an interesting asymmetry. Men tend to produce more humor attempts in flirtatious interaction; women tend to respond to humor more (laughing more). This pattern has been interpreted through multiple lenses: the evolutionary view that humor production signals cognitive fitness and therefore functions as a male display; the socialization view that cultural norms around gender and comedy shape who is expected to be funny and who is expected to be appreciative. What is clear is that shared laughter — both parties finding the same thing genuinely funny, rather than one party performing jokes and the other performing appreciation — is a strong positive signal in early interaction. The experience of genuinely surprising someone into laughter, and being surprised into laughter yourself, creates a moment of authentic shared reaction that is difficult to manufacture and therefore carries real information about compatibility.

📊 Research Spotlight: Bressler, Martin, and Balshine (2006) studied humor appreciation and humor production in mate preferences and found that women prioritized having a partner who made them laugh (receptive appreciation of their own humor) while men prioritized having a partner who laughed at their humor. If replicated, this is a fascinating asymmetry — both sexes want humor in a partner, but they want different aspects of it. More recent research has partially complicated this finding by showing that the effect depends heavily on the level of relationship seriousness being considered, but the basic asymmetry has shown some robustness.

19.10 Digital Flirtation: Memes, Voice Notes, and New Media Forms

Digital communication has not eliminated flirtation — it has created new registers and new constraints for its performance. Understanding how flirtation maps onto digital media requires attending to the specific affordances and limitations of each platform.

Text-based flirtation (SMS, DMs, WhatsApp chat) strips out several of the major nonverbal channels — gaze, proximity, touch, and much of vocal prosody are unavailable. What remains is linguistic content and its timing. Research on digital self-presentation in flirtatious contexts (extending the work we will examine more fully in Chapter 20) finds that:

• Response latency takes on enormous signal weight. How long someone takes to reply, and whether the response time accelerates over a conversation, is read as an indicator of interest or its absence.
• Message length and linguistic elaboration are proxies for effort and investment
• Emoji and reaction use — including specific emoji sequences and the layering of reactions on others' messages — have developed into a rich paralinguistic system that partially replaces nonverbal channels
• Typos and deliberate informality can signal casual intimacy; excessive correction can signal social distance

Voice notes are an interesting hybrid form: they restore vocal prosody (pace, pitch, breathiness, rhythm) while remaining asynchronous and composable. Some research suggests that voice notes in early-relationship flirtation carry more emotional information than text and are perceived as more intimate — using them signals a willingness to be heard, not just read.

Memes and reactions as flirtatious signals have received less formal research attention, but represent a genuinely novel form of courtship communication. Sending someone a meme is a low-stakes act of attention-paying — it says "I saw this and thought of you" — while the content and specifically the act of sending can be calibrated to signal varying levels of interest. It is, in structural terms, the digital equivalent of a low-stakes touch: a small bid for connection that can be accepted, reciprocated, or ignored without either party losing face.

The paradox of digital clarity and digital ambiguity is worth noting: text communication is in some ways more careful and crafted than live conversation — you have time to compose, revise, and present an edited version of yourself. But the removal of nonverbal channels does not reduce ambiguity; it redistributes it. Without vocal tone and facial expression to constrain interpretation, words carry more interpretive latitude. The same message can be read as warm, ironic, or disinterested depending on assumptions the receiver brings to it. Digital flirtation is, if anything, more ambiguous than face-to-face flirtation in many respects — it just feels more manageable because you have time to think.

19.11 Cross-Cultural Flirtation: Does It Translate?

Reyes's behavioral coding team, working in Sweden and Germany, documented a pattern they called "low-intensity maintained contact": prolonged conversations at moderate volume, with subdued gesture, sustained mutual gaze, and minimal physical touching until later in the interaction. Okafor's team in Nigeria coded what they termed "high-expressiveness bidding": elevated gesture, louder vocal exchange, more physical proximity earlier in the interaction, and more explicit verbal play (direct compliments, open teasing).

Both patterns, analyzed against local norms, served the same structural function: they were the locally appropriate way to signal interest while maintaining the deniability frame. The Swedish interaction looked "reserved" by Nigerian norms; the Nigerian interaction looked "forward" by Swedish norms. But in context, both were recognizable to local participants as flirtation rather than as simple friendliness.

This finding — structural universals with substantial content variation — is consistent with the theoretical framework of Section 19.7. If the function of flirtation is to signal interest within a deniable frame, the specific behaviors used to accomplish this are culturally negotiated. Any behavior that the local culture treats as simultaneously signaling interest and retaining deniability can serve this function. The repertoire is culturally specific; the architecture is not.

Cross-cultural flirtation encounters — when people from different flirtation cultures interact — create predictable misreadings in both directions. Someone from a high-expressiveness flirtation culture may read a subdued interaction partner as uninterested when they are in fact interested but performing interest in low-expressiveness register. Someone from a low-expressiveness culture may read high-expressiveness behavior as romantic interest when it is simply the normal register of social friendliness.

The South African data added an important dimension to this picture. Okafor's South African coders documented significant variation within South Africa by language community and urban versus township context — suggesting that even within a single country, flirtation behaviors are not uniform. The relevant cultural unit for understanding flirtation is not the nation-state but the more specific subculture of socialization.

🧪 Methodology Note: Cross-cultural behavioral coding faces significant challenges. The behaviors that count as flirtatious in one cultural context may not have any natural mapping to behaviors in another. Okafor and Reyes spent nearly eight months iterating on their coding protocol to develop a system that was neither ethnocentrically derived from Western norms nor so generic as to be uncodeable. The solution they settled on — coding function rather than specific behavior, using local informants to validate interpretations — represents a methodological advance that will be discussed in Chapter 22.

19.11b Intersectionality and Cross-Cultural Flirtation: Further Complications

The Okafor-Reyes data, as the previous section described, documented variation in flirtation surface structure across cultural contexts. But within those already-varied cultural contexts, further intersectional variation exists that the macro-level cross-cultural comparison can obscure.

Consider gender and sexuality. The behavioral patterns that Moore's ethological work documented — female solicitation, male approach — were observed in predominantly heterosexual interaction in Western settings. LGBTQ+ courtship contexts operate with different structural features, because the gender-binary organization of who initiates which phase does not apply in the same way. Research by Gonzaga and colleagues on same-sex attraction and courtship found that the early phases of same-sex flirtation often involve more explicit symmetry — both parties signaling and both parties approaching — than the heterosexual initiation literature suggests is typical. This makes sense structurally: if neither party is assigned the "solicitor" role and neither the "approacher" role by gender norms, both parties must be more explicit about their availability.

Race also intersects with flirtation in ways that deserve acknowledgment. The racialization of desirability — the well-documented pattern in which racial and ethnic groups experience systematically different rates of desired and undesired attention in dating contexts — shapes the flirtation experience in ways beyond what behavioral coding can capture. Being a member of a group that is either hyper-sexualized (fetishized) or desexualized (ignored) in the dominant cultural imagination shapes how flirtatious signals are sent and received in ways that the neutral behavioral coding framework cannot fully account for. This is part of what the replication crisis section of this book asks us to hold in mind: the populations in which research is conducted shape the findings, and a science of flirtation built on observations of predominantly White, heterosexual, Western adults has real limitations.

🔵 Ethical Lens: The behavioral coding tradition in courtship research — including both Moore's landmark work and the Okafor-Reyes protocol — operationalizes flirtation through observable behavior without necessarily capturing the subjective experience of the people involved. A behavior that is coded as a solicitation signal by an observer may be experienced by the performer as routine friendliness. A behavior coded as neutral may be experienced by the target as unwanted. The gap between behavioral observation and subjective experience is always present in social science; in courtship research, it is especially consequential.

19.12 The "Friend Zone" as Misread Flirtation: What the Research Says

The "friend zone" — the popular conception of a situation where one person wants more than friendship while the other wants only friendship — has generated both enormous amounts of popular commentary and a modest but interesting body of research. What does the literature actually find?

The core finding, replicated across several studies, involves cross-sex friendship perception asymmetry: when men and women who are friends report on their feelings toward each other and their perceptions of the other's feelings, systematic discrepancies emerge. Bleske-Rechette and Buss (2001) and subsequent researchers found that men tend to overestimate women friends' romantic interest in them, and that men are more likely to be romantically interested in female friends than women are in male friends.

This cross-sex perception asymmetry appears related to the same sexual overperception bias discussed in Section 19.8. Friendly behavior, including laughter, warmth, sustained attention, and physical closeness — many of the same behaviors that function as flirtatious signals in one context — also appear in genuine friendship. The ambiguity that is structural to flirtation makes its signals interpretable as either flirtatious or friendly, and who makes which interpretation appears to be partly a function of the observer's own level of interest.

It is important to be clear about what this research does and does not show:

It does not show: That women routinely "lead men on" or behave deceptively in cross-sex friendships. The asymmetry in perception is real; the source of deception (if any) is not in the behavior but in the interpretation.

It does show: That friendliness and flirtation use overlapping behavioral repertoires, and that when one party is romantically interested, they are prone to interpreting the other's friendliness through a romantic frame that the behavior does not necessarily support.

The intersectional dimension: Research on cross-sex friendship perception has been conducted almost entirely with heterosexual participants. The dynamics in LGBTQ+ friendships, or in friendships that span orientation categories, are both theoretically different and understudied. Jordan, as a nonbinary queer person navigating relationships across gender and orientation, would recognize that the "friend zone" concept is organized around a heteronormative binary that poorly describes many actual relationship configurations.

🔴 Myth Busted: The popular claim that the "friend zone" is something women do to men — a kind of relational cruelty or manipulation — is not supported by the research. The asymmetry in romantic interest and perception is real, but it arises from the interpretive tendency of the interested party, not from deceptive behavior by the less-interested one. Friendly behavior is not a promise of romantic interest, and interpreting it as such reflects the interpreter's desires more than the friend's signals.

19.12b The Replication Crisis and Flirtation Research

This textbook has emphasized throughout (see especially Chapters 3 and 12) the importance of methodological humility and awareness of the replication crisis in psychology. How does this apply to the flirtation research reviewed in this chapter?

The honest answer is: unevenly. Some findings are more robustly supported than others.

Better-supported findings: - The existence of a sequential structure in courtship interaction (Perper, Grammer, and subsequent observational work) - The role of behavioral reciprocation in courtship escalation - The sexual overperception bias as a population tendency (replicated in multiple samples) - The prevalence and functional importance of behavioral ambiguity in flirtation (theoretically convergent across multiple frameworks) - The existence of cross-cultural variation in specific flirtation behaviors alongside structural similarities

Less-robustly supported findings: - Specific claims about the exact behavioral content of Moore's 1985 catalog (the sample was culturally specific and the behaviors may not generalize) - The precise evolutionary mechanisms proposed for sex differences in courtship initiation - Claims about the specific neural or hormonal substrates of flirtation (research is sparse and highly subject to WEIRD-sample limitations) - The sharp distinction between "terminal" and "recreational" flirtation as distinct categories (this may be more of a spectrum than a binary)

Understudied areas that should make us cautious about generalizing: - LGBTQ+ flirtation - Flirtation in non-Western, non-industrialized contexts - Flirtation across age groups outside college student samples - Digital flirtation (a genuinely new domain with very limited longitudinal data)

🧪 Methodology Note: The behavioral coding tradition in flirtation research faces an inherent interpretive challenge: the researchers who code behavior must make judgments about whether a given behavior counts as a courtship signal. These judgments are inevitably shaped by the researchers' cultural assumptions, theoretical frameworks, and observer expectations. The inter-rater reliability statistics reported in studies like Moore's tell us how consistently two coders agreed — but they do not tell us whether both coders were making the same culturally-embedded error. This is a limitation that more recent cross-cultural work, like the Okafor-Reyes protocol, has attempted to address through multi-site, multi-culture teams with culturally-insider coders. But it is not fully solved.

19.13 Returning to Okafor and Reyes: What the Data Settled and What It Didn't

After their data comparison, Okafor and Reyes wrote a joint position paper for the Year 3 report of the Global Attraction Project. In it, they proposed what they called the "dual-structure" model of flirtation: that flirtation has a universal deep structure (graduated, reciprocation-dependent signaling within a deniable frame) and a culturally specific surface structure (the particular behaviors that instantiate this structure in any given cultural context).

Reyes was satisfied that this vindicated what he had been arguing about biological priors — the deep structure, he maintained, was shaped by evolutionary pressures that transcended cultural variation. Okafor was satisfied that it vindicated her skepticism about assuming any specific behavior was universal — the content variation, she argued, was as theoretically important as the structural commonality.

What the data could not settle was the question of mechanism: whether the deep structure was there because of selection pressure, cultural convergence, or some combination. That question, both acknowledged, was not yet answerable. But the empirical finding — structural universality with content variation — was, they both agreed, the most interesting result the Year 3 fieldwork had produced.

The practical implications for students are clear: when you move across cultural contexts, do not assume your flirtation vocabulary is shared. The function is shared; the form is not. And when you are navigating flirtation in your own cultural context, the most important thing to understand is that the ambiguity you experience is not noise in the signal — it is the signal's architecture.

19.13b What the Data Did Not Settle: The Mechanism Question

Okafor and Reyes's joint position paper was published in the Year 3 report with unusual scholarly candor: a section titled "What We Cannot Yet Conclude." In it, they acknowledged that the dual-structure model described a pattern in the data without explaining the mechanism that produces it.

Is the structural universality of flirtation a product of evolutionary selection pressure — a set of behavioral tendencies shaped over evolutionary time because they solved recurrent coordination problems in human mating? Or is it a product of cultural convergence — independent cultural inventions of a similar solution to the social problem of courtship face risk, which happens to produce similar structural results across cultures without any shared evolutionary cause? Or is it some combination: biological tendencies that shape the architecture of possible cultural scripts, with culture filling in the specific content?

Reyes favored the first interpretation; Okafor favored the second; the data did not adjudicate between them. This is not a failure of the study — it is an honest representation of where the scientific frontier lies. The behavioral data can establish what flirtation looks like and how it is structured. It cannot, on its own, establish why the structure exists. That is a question for converging evidence from multiple methods: comparative animal behavior, neuroimaging, developmental psychology, cross-cultural variation studies, and historical analysis of how courtship scripts have changed over time.

For students, the lesson is important: in psychology and social science, establishing a pattern and explaining a pattern are different tasks. The dual-structure finding is a real and interesting empirical result. Its explanation is, honestly, still open.

19.14 Summary: The Architecture of Flirtation

This chapter has analyzed flirtation as a social performance with several interlocking properties: it is culturally scripted yet individually improvised; it is strategically ambiguous yet structurally recognizable; it is functionally rational yet experientially playful. These are not contradictions. They are the features of a sophisticated social behavior that has evolved to accomplish multiple things simultaneously.

We have covered:

What flirtation is: Indirect, playful, deniable signaling of potential romantic interest within a friendly-sociability frame — with the triangulation challenge of aligning behavioral, intentional, and perceptual levels of analysis.

The dramaturgical framework: Goffman's concepts of front stage/back stage behavior, frame, and face work provide the most elegant conceptual vocabulary for understanding why flirtation has the structural properties it does.

Flirtation scripts and improvisation: The Perper-Moore sequence documents a real sequential structure in courtship interaction that functions like a jazz standard — a shared framework within which individuals improvise. Script violations in either direction (too fast, too slow) carry social costs.

Female initiation and male approach: Ethological research, especially Moore (1985) and Grammer (1990), documents that the initiation of heterosexual courtship is more often female (through solicitation signals) than the dominant cultural narrative suggests, with male approach following female invitation.

Male display behaviors: Men's courtship display behaviors — resource signaling, vocal modification, humor production — function primarily in the approach and self-presentation phases.

The ambiguity function: The most theoretically rich finding in the chapter. Flirtation's ambiguity is its design, serving face-protection, quality-testing, arousal-maintenance, and strategic-coordination functions simultaneously. The game-theoretic analysis shows how a sequence of low-cost, deniable signals can nonetheless be informationally meaningful when taken together.

Rejection ambiguity: The sexual overperception bias is a real, replicated population tendency that has real consequences. Ambiguity in rejection signals is partly a product of the same face-protection logic that drives flirtatious ambiguity itself.

Playfulness, laughter, and humor: Laughter and teasing maintain the non-serious frame while building intimacy; humor production and appreciation show an interesting cross-sex asymmetry in what partners want.

Digital flirtation: New media forms have not eliminated flirtation's structural logic but have redistributed it across new channels, with response latency, message elaboration, and emoji use substituting for unavailable nonverbal signals.

Cross-cultural variation: The Okafor-Reyes dual-structure model — universal deep structure, culturally variable surface content — represents the most empirically grounded synthesis of universalism and cultural specificity currently available.

The "friend zone": The cross-sex perception asymmetry is real; the attribution of its source to deceptive behavior by the less-interested party is not supported. The misread is in the interested party's interpretation, not in the friend's behavior.

The chapter's central claim stands: flirtation is a sophisticated social technology that has been refined to accomplish difficult social goals — signaling interest while managing risk, proposing intimacy while maintaining deniability, creating connection while preserving freedom. Understanding it as a performance, with a script and improvisational space within that script, is more accurate and more respectful of its genuine complexity than treating it as either a transparent emotional expression or a deliberately manipulative act.

In Chapter 20, we turn to digital platforms and examine how the structural logic of courtship and flirtation adapts — and what changes fundamentally — in algorithmic, app-mediated environments.

Chapter 20 preview: Swiping in the Dark — Dating Apps, Algorithms, and the Gamification of Desire